ULTRABENCH Created by Iternal Technologies, the AI Strategy Experts

Intelligence vs Cost

Every ranked LLM plotted by intelligence index against blended price per 1M tokens, with the best index per dollar ranked beneath.

Best index per dollar
ModelIndexBlended $/MIndex / $
Ling 3.0 Flash inclusionAI66.4$0.03152,107.6
GLM 5.3 Flash Zhipu AI (Z.ai)78.8$0.04951,590.3
gpt-oss-20b OpenAI46.8$0.0361,299.4
Muse Spark 1.3 Contributor Meta87.1$0.125696.6
DeepSeek-V4-Flash DeepSeek76.9$0.1125683.6
Ling 3.0 Flash VL inclusionAI53.4$0.09593.3
Ripple 65B Preview Ricursive81.6$0.15543.7
Gemma 4 26B-A4B Google DeepMind60.8$0.1211502.2
Granite 4.2 8B IBM41.3$0.0825500
Qwen3.5-Flash superseded Alibaba56.4$0.1138496.2

Index ÷ blended $/1M tokens, top 10. Models below index 40 are excluded — a cheap model that answers nothing is not value. The board's Index / $ column ranks every priced model, with no floor.